As the U.S. economy slows, the question naturally arises: are we in a recession? The answer is more complicated than it appears for a number of reasons. First, the data economists use to assess the economy are available with a lag of at least one month. Second, the data are often subject to revision in subsequent months and years and these revisions can change the assessment significantly. Third, once the economy reaches a turning point (peak or trough), it takes at least a few months for that to become apparent in the data. These factors combine to make identifying the onset of a recession using economic data a backward-looking exercise.
While some analysts define recessions as at least two consecutive quarters of negative real GDP growth, for the U.S. most economists rely on recessions identified by the National Bureau of Economic Research (NBER). The NBER is a private, nonpartisan organization that facilitates cutting-edge investigation and analysis of major economic issues.
Within the NBER, a group of economists serve on the Business Cycle Dating Committee. They have identified monthly peaks and troughs in the U.S. economy going back to the mid-1800s. They define a recession as a sustained, generalized, and significant decline in the level of economic activity. Slowing growth is not enough. Economic activity must have declined.
As part of this definition, the committee looks for downturns of a certain depth, diffusion, and duration. In other words, the downturn must be deep enough, present across a wide enough range of markets and sectors, and last long enough. In general, they look for downturns at least as deep, diffused, and long as the mildest prior recession. There are no rigid rules to this. Sometimes one of the criteria is not strictly enforced.
A recession begins when the economy reaches a peak and subsequently declines. It ends when the economy reaches a trough and subsequently expands. Similarly, an expansion begins when the economy reaches a trough and ends when the economy reaches a peak.
On average since World War II, recessions have lasted 10 months and expansions have lasted 64 months. The shortest post-WWII downturn lasted two months (February 2020 to April 2020) and the longest lasted 18 months (December 2007 to June 2009). The shortest expansion lasted 12 months (July 1980 to July 1981) and the longest lasted 128 months (June 2009 to February 2020).
The NBER committee looks at a wide range of indicators, but two of the most important are real personal income less transfer receipts and nonfarm payroll jobs. In addition, they look at total employment from the household survey, industrial production, real personal consumption expenditures, and wholesale-retail sales adjusted for price changes. Many other indicators are considered as well, including real GDP.
Exhibits 1-6 show the recent history for six of the indicators. Most indicators include data through June 2022. The exception is real manufacturing and trade industries sales which is through May 2022.
Let’s assume that we need five or six months of data after any prospective date to determine that a series has peaked and begun to decline. Clearly, U.S. nonfarm payroll jobs did not peak in January 2022. Likewise, real personal income less transfers continued to unevenly increase through June of 2022.
Real personal consumption expenditures, household survey employment, and industrial production all also continued to increase from their January 2022 level.
For the manufacturing and trade sales data, should we identify a peak in March 2021 or even January 2022? Maybe so. If sales continue to drop in 2022, the peak for this series will likely be placed on one of those months.
Based on currently available data, it is difficult to claim that the U.S. economy entered a recession in January 2022. Further, it seems unlikely that the economy entered a recession in February or March either.
The data for April, May, and June show significantly slower growth for all indicators except jobs. However, we do not yet have enough data to confidently identify a peak in any series except real manufacturing and trade industries sales.
Exhibit 1: U.S. Real Personal Income Less Transfers, Seasonally Adjusted, Billions of Chained 2012 Dollars
Exhibit 2: U.S. Total Nonfarm Payroll Jobs, Seasonally Adjusted
Exhibit 3: U.S. Household Employment, Seasonally Adjusted
Exhibit 4: U.S. Total Industrial Production, Seasonally Adjusted
Exhibit 5: U.S. Real Personal Consumption Expenditures, Billions of Chained 2012 Dollars
Exhibit 6: U.S. Real Manufacturing and Trade Industries Sales, Millions of Chained 1996 Dollars
George W. Hammond, Ph.D., is the director and research professor at the Economic and Business Research Center (EBRC).